Niobium Skills
Build apps that compute on data no one can see, not even the server running them. Agent skills that guide your AI coding agents from privacy idea to a working encrypted…
Install: /plugin marketplace add NiobiumInc/niobium-skills
Works with Claude Code · 14 GitHub stars
Install
/plugin marketplace add NiobiumInc/niobium-skillsInstall command taken from the project's README or registry entry. Always review code before running it.
README excerpt
Read full README on GitHub →Niobium Skills
AI Agent Skills by Niobium — portable SKILL.md
packages that guide AI coding agents through building privacy-preserving and fully
homomorphic encryption (FHE) applications. Each skill works with any agent that supports
the open Agent Skills standard: Claude Code, OpenAI Codex CLI, Cursor, Gemini CLI, GitHub
Copilot, Windsurf, and more.
Skills
fhe-application-design — design FHE applications (a.k.a. FHEanna, "fee-AH-na" · patent pending)
An eleven-stage methodology (Stages 0–10) for designing and building Fully Homomorphic Encryption (FHE) applications with OpenFHE (optional Niobium DSL path) — from build environment and privacy model through feasibility, scheme selection, circuit design, parameter selection, the faithful twin, encrypted implementation, protocol specification, and a Niobium Fog deployment. Use it whenever you need to compute on encrypted data, assess FHE feasibility, or structure a client–server protocol for encrypted computation.
npx skills add NiobiumInc/niobium-skills --skill fhe-application-design
Using Claude Cowork? Don't use the CLI above — it targets coding agents (Claude Code) and its
--globaloption won't apply to Cowork. Install through Cowork's plugin UI instead: see Claude Cowork below.
Prerequisites (beyond the skill itself): building and running the encrypted
application (Stages 8 and 10) uses the FHE-dev container, which ships
Niobium's instrumented OpenFHE fork + libnbfhetch (built from
niobium-client). You need Docker.
Pull the prebuilt image (recommended), or build it yourself from the skill's
Dockerfile (the first build compiles OpenFHE from source):
# Pull the prebuilt image (recommended)
docker pull ghcr.io/niobiuminc/fhe-dev:latest
docker run --rm ghcr.io/niobiuminc/fhe-dev:latest make test-release # smoke test
# Or build it yourself (compiles OpenFHE from source on the first build)
docker build -t ghcr.io/niobiuminc/fhe-dev:latest skills/fhe-application-design/environment
The skill walks you through this as its Stage 0, so you can also just start a conversation and let the agent set it up with you. The design stages (1–7) need nothing installed at all.
See the skill's README for full documentation.
| Skill | What it does | Install |
|---|---|---|
fhe-application-design |
Design & build FHE applications (OpenFHE / optional Niobium DSL) | npx skills add NiobiumInc/niobium-skills --skill fhe-application-design |
Installation
Each skill installs independently. Pick whichever method fits your setup — all of them
install the full skill folder, including its references/.
Universal CLI (recommended)
npx skills add NiobiumInc/niobium-skills --skill <skill-name>
skills.sh auto-detects your agent and writes the skill to the right directory.
Claude Cowork
Cowork installs skills as plugins through its own UI — not the npx/gh
CLIs above (those write to a coding agent's .claude/skills/, which is why they
report a "Claude Code" target and offer a --global option that doesn't apply
to Cowork). Add this repo as a plugin marketplace, then install the plugin:
Similar plugs
Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
An AI skill that provides design intelligence for building professional UI/UX across multiple platforms.
Turn any codebase, with its docs, SQL schemas, configs, and PDFs, into a queryable knowledge graph. A /graphify skill for Claude Code, Cursor, Codex, and Gemini CLI: local…
Why use many token when few token do trick. Viral skill + proxy for coding agents that cuts 65% of tokens by talking like a caveman.
Production-grade engineering skills for AI coding agents.